Research On Optimal Control Technology Of High Resolution Imaging System Based On Adaptive Optics And Image Processing

For high-resolution imaging systems,especially for large-diameter ground-based telescopes,the disturbance of external distorted wavefront and the aberration inside the optical system are the main factors that hinder high-resolution imaging.The main solutions are adaptive optics technology,post image restoration technology and hybrid processing technology(adaptive optical correction + post image processing).When the first two technologies are used alone,the ability to correct aberrations is limited.At present,hybrid processing technology is the mainstream method.However,the essence of this "hybrid" is the simple splicing of adaptive optics technology and image processing technology.The control method of the wavefront corrector still uses the traditional idea of minimizing the RMS of residual error,resulting in the output of the system-the restored image-is not constrained by the control method.Therefore,for the hybrid processing system,it is of great significance to study a control method that combines the two and aims to improve the quality of the final restored image.In order to optimize the quality of the image obtained by the imaging system,this dissertation studies the control method based on the combination of adaptive optics technology and image restoration technology,as well as the new control method of adaptive optics system,and obtains some innovative conclusions and achievements.1.Based on the theoretical analysis of the traditional hybrid processing system(phase-conjugate adaptive optical correction + image post-deconvolution),it is concluded that the quality of the restored image is closely related to the performance of the wavefront corrector and the wavefront detector.Zernike model is used to analyze the spatial fitting error of deformable mirror.The results show that it can introduce a large number of high-order derivative aberrations while fitting low-order aberrations.The Hartmann wavefront sensor is equivalent to a low-pass filter,so the characteristics of correction residual will reduce the accuracy of wavefront restoration,resulting in the distortion of restored image.Based on the description of the surface displacement of deformed mirror,this dissertation puts forward the concept of correction degree(the scaling ratio of the control voltage of the deformable mirror to that of the traditional method).According to this concept,an optimal control method is established,which can find the best balance between the wavefront reconstruction precision and the RMS of total residual error.Theoretical analysis and experimentalresults show that this method can improve the quality of restored image.2.The traditional optimized adaptive optical system usually lacks post image restoration,which leads to the image quality cannot be further improved.Based on the traditional optimized adaptive optical system,this dissertation proposes two hybrid processing systems combined with image restoration technology,and both of them adopt the control method based on stochastic parallel gradient descent(SPGD)control algorithm.In the first hybrid processing system,the adaptive optical wavefront correction part and the image restoration part are relatively independent.With the help of wavefront detection information,the deformable mirror control signal is calculated and used as the initial value of the iterative algorithm.This prior knowledge can greatly accelerate the convergence speed of the algorithm.After the iteration,according to the information of the wavefront detector and the far-field image,the post Wiener deconvolution is carried out to get the final restored image.In the second hybrid processing system,image restoration is put into the iterative loop of SPGD algorithm as a whole.The performance metric of the algorithm is the quality function of the restored image.After the iterative convergence,the quality of the restored image is also optimal.Theoretical analysis and experimental results show that,compared with the traditional methods,the two methods proposed in this dissertation have significantly improved the correction ability under different turbulence conditions,and show excellent robustness.3.Some special Zernike modes have interrelated relationship in the concentric aperture circle domain of unit circle.When the modes with strong negative correlation is superposed with a certain coefficient,the aberrations in a certain concentric aperture will cancel each other.This phenomenon is called the conjugate property between modes.Inspired by this,the correlation between the low-order and high-order parts of the residual wavefront aberration for adaptive optical system is analyzed.The results show that in the residual wavefront aberration with large RMS,there is a significant negative correlation between the two parts of aberrations,and the two parts of aberrations will show conjugate property in a certain combination.Based on this conclusion,the author designs an optimal control method,which searches for the best correction degree to achieve the best coefficients of low-order and high-order aberrations.In the optimal case,the RMS of the residual wavefront in a certain concentric aperture will be reduced to the extreme value,which can improve the imaging quality in this aperture range.Theoretical analysis and experimental resultsshow that this method can improve the imaging ability of the adaptive optical system in the face of complex aberrations,and effectively expand the scope of application of the traditional adaptive optical system.In this dissertation,several optimal control techniques of high-resolution imaging system are proposed,which have important application value in improving the optical imaging quality of adaptive optical system and the restoration image quality of hybrid processing system.In particular,it has an important reference significance in improving the overall performance of the hybrid processing system.